The University of Texas System

Other

Khosrow Behbehani

John R Burk

Edgar A Lucas

Fu-Chung Yen

Sridhar Vijendra

John R Axe

Michael T Manry

Francisco J Lopez

Mohammad A Al-Abed

Edgar R Lucas

Described herein is a method of developing a fuzzy logic system to detect a non-normal health condition. In particular, signal processing and transformation of electrocardiogram (EKG) signals for sleep disorder breathing are provided. The method includes: recording EKG measurements during sleep; clipping the measurements into clips of a consistent length; calculating heart rate and obtaining an evenly sampled discrete time series data clip; performing Short-Time Discrete Fourier Transform on each data clip generating STDFT respective matrices; encoding each STDFT matrix into a grey-level image; calculating Grey-Level Co-occurrence Matrices; extracting textural features; performing statistical analysis on the features to formulate rules; and employing the rules in a Fuzzy Logic system. The system and method described herein yields an accuracy of 75.88%, or better, in detection of sleep apnea.

Processing and analyzing physiological signals to detect a health condition

A system to form and store an electrocardiogram (ECG) signal derived from a cardiac electrical signal that includes an apparatus having a pair of the electrodes to connect to a patient to detect the cardiac electrical signal. A signal sampler samples the cardiac electrical signal to form the ECG signal. A data storage device stores the ECG signal. A computer communicates with the data storage device to retrieve the ECG signal for analysis by software stored in the memory of the computer. The software analyzes a morphology of the amplitude of a plurality of R-wave peaks contained within the ECG signal and/or analyzes a morphology of the area of a plurality of QRS complex pulses contained within the ECG signal.

System, software, and method for detection of sleep-disordered breathing using an electrocardiogram

A pressurized breathable gas, preferably air, is supplied to the patient's airways with an interface including a mask and a flexible hose. A selected oscillation component or forced probing signal is applied or superposed on the pressurized breathable gas, preferably with a loudspeaker coupled to a frequency generator. Pressure and flow of the breathable gas in the interface are measured or sampled, preferably using pressure transducers and a pneumotachometer coupled to a personal computer, to obtain pressure and flow data characteristic of the patient's airway. The pressure and flow data are converted to the frequency domain using Fourier transform or cross-power spectrum techniques in the personal computer. A function of the processed pressure and flow data is obtained to determine an index of airway impedance or degree of airway obstruction.

Method and apparatus for detection and diagnosis of airway obstruction

A method and apparatus for treating sleep disorder breathing is disclosed having improved ability to accurately detect pharyngeal wall vibration or other apneic events. An interface or mask is placed over a patient's airway. The interface is coupled to a source of pressurized gas. A respiration-related variable, namely the total pressure in the interface, is measured or sampled. The respiration-related variable is input into an artificial neural network trained to recognize patterns characterizing sleep disorder breathing. Responsive to recognition by the artificial neural network of sleep disorder breathing, pressurized gas is supplied to the patient's airways through the interface.

Method and apparatus for treatment of sleep disorder breathing employing

A method and device for controlling sleep disorder breathing utilizes variable pressure. The compressor supplies air at a relatively low pressure to the user's air passages while the user is asleep. A pressure transducer will monitor the pressure and convert the pressure into an electrical signal. The electrical signal is filtered and processed to compare it to the characteristics of waveforms that exists during snoring. If the envelope of the waveform exceeds an average threshold value in duration and in area, then the microprocessor will consider the envelope possibly associated with a snore. If a selected number of envelopes of this nature occur within a selected time period, then the microprocessor considers snoring to exist and increases the pressure of the compressor. If snoring is not detected within a certain time period, then the microprocessor lowers the level gradually.

Growing community of inventors

Arlington, TX, United States of America

Khosrow Behbehani